ZeroClaw is an open-source, local-first AI agent and automation platform. You can use ZeroClaw to deeply integrate multi-channel communication capabilities with large language models (LLMs) to create custom AI assistants with persistent memory and proactive execution capabilities.
ZeroClaw is an open-source AI assistant. Before you use it, fully assess its security and stability and strictly follow the license agreement to protect your system and data.
Billing
Using the ZeroClaw service on a Simple Application Server incurs the following costs:
Simple Application Server fees: See Simple Application Server Billing overview.
Model invocation fees: By default, ZeroClaw uses a Model Studio model. There are two primary billing methods for the model:
Coding Plan (recommended): This plan has a fixed monthly fee for a set monthly request quota. Requests that exceed the quota are blocked and not billed, preventing unexpected costs.
The Coding Plan currently supports models such as
qwen3.5-plus,kimi-k2.5,MiniMax-M2.5, andglm-5.Pay-as-you-go: By default, the
qwen3.5-plusmodel is used for text and image processing.ImportantAutomatically suspend service to avoid charges: When you first activate Model Studio , you automatically receive a free quota for each model. After the free quota is used, billing automatically switches to pay-as-you-go. To prevent additional charges, we recommend enabling the feature to suspend the service when the free quota is exhausted.
Step 1: Purchase the ZeroClaw package
If you already have a Simple Application Server instance, you can skip this step and deploy ZeroClaw by resetting the system. To do this, select the ZeroClaw image from the application images. However, resetting the system erases all data on the system disk. Back up your data as needed. For detailed instructions, see Reset system.
Purchase a Simple Application Server with ZeroClaw pre-installed.
On the Simple Application Server page, the resource configuration is preset. If you need to change any settings, refer to the following guide:
Servers: The instance must have at least 2 GiB of memory. You can switch plans based on your requirements.
Region: The default is China (Beijing).
Purchase Settings: The default subscription is for 12 months. You can change the duration as needed.
Step 2: Configure ZeroClaw
In the Simple Application Server console, click the instance ID on the server card. On the Server Overview page, click the Application Details tab.
In the ZeroClaw Usage Steps area, under Allow Port Access, click One-Click Enable to open the required port in the firewall for the ZeroClaw service.
ImportantTo prevent malicious scans and targeted attacks, ZeroClaw does not use a fixed default port. Instead, it automatically generates a random port during initialization. You can view the ZeroClaw port number in the console.
Under Configure ZeroClaw, click Initialize to configure your Model Studio API key.
You can configure two types of Model Studio API keys:
Coding Plan exclusive API key (recommended): This plan has a fixed monthly fee for a set monthly request quota. Requests that exceed the quota are blocked and not billed, preventing unexpected costs.
The Coding Plan currently supports models such as
qwen3.5-plus,kimi-k2.5,MiniMax-M2.5, andglm-5.A pay-as-you-go Model Studio API key.
You can configure the API key by using either the System Recommendation or Manual Input method. The system recommends a cost-effective API key from your Coding Plan and a low-latency API key for the Model Studio service access point nearest to your server. If you need to use an API key from a different region or account, choose Manual Input.
System Recommendation (select from the dropdown list)
After making your selection, click Next.
Server region
Recommended API region
Coding Plan region
Chinese mainland regions
China (Beijing)
China (Beijing)
US and European regions
US (Virginia)
Hong Kong (China) and other Asian regions
Singapore
Manual Input
Click the button to switch to manual input. Enter the Model Studio API key, select its corresponding region, and click Next.
Note that this step overwrites any existing ZeroClaw configurations for the Model Studio API key, service access point, and model. The system displays the optimal Model Studio API key based on your instance's region and automatically configures the nearest Model Studio service access point.
ImportantWhen you manually configure the API key, ensure that you select the correct region. An incorrect region prevents proper model invocation.
Access the WebUI panel.
The pairing code is valid for one-time use and is bound to the current browser session. Opening the URL in a new browser tab or window requires a new pairing code.
In the Generate Pairing Code section under Access WebUI Panel, click Run Command to get the pairing code for the ZeroClaw WebUI.
In the Get WebUI Panel Address section under Access Web UI Panel, click Run Command to get the web panel URL. Open the URL and enter the pairing code to start using the ZeroClaw agent.
(Optional) Integrate message channels. You can configure message channels such as DingTalk, Lark, and QQ for ZeroClaw.
FAQ
View the pairing code
Navigate to and click Run Command to view the pairing code.
Do not share your pairing code. The pairing code is the sole authentication credential for the ZeroClaw gateway. Anyone who has the code can pair with the gateway and gain full control.
Change the model
Go to the WebUI and click Configuration in the left-side navigation pane.
In the General section, update the following fields:
API key: Your model's API key.
API URL: The base URL for the model's service.
Default Provider:
custom: <model_base_URL>.Default Model: The name of the model you want to use.
Click Save in the upper-right corner to save your settings.
Restart the ZeroClaw gateway
If you experience a connection loss or service unavailability, you can restart the ZeroClaw gateway from the console. Navigate to and click Run Command.
View the ZeroClaw port number
To prevent malicious scans and targeted attacks, ZeroClaw generates a random port on initialization instead of using a fixed default. To view the port number, navigate to and click Run Command.
API key not recognized
The system detects the nearest service access point based on your server's region and displays the corresponding API key. It also shows the API key for your Coding Plan, if you have one. If an API key for a nearby service access point does not exist, follow the on-screen prompts to create one or to subscribe to a Coding Plan.
Server region | Recommended API region | Coding Plan region |
Chinese mainland regions | China (Beijing) | China (Beijing) |
US and European regions | US (Virginia) | |
Hong Kong (China) and other Asian regions | Singapore |